BigDL
BigDL (https://bigdl-project.github.io/0.6.0/) is an open source, distributed, deep learning framework for Apache Spark implemented by Intel (https://www.intel.com). It is licensed with the Apache 2.0 license, the same as for DL4J. It has been implemented in Scala and exposes APIs for Scala and Python. It doesn't provide support for CUDA. While DL4J allows cross-platform execution in standalone mode (including Android mobile devices) and distributed mode (with and without Spark), BigDL has been designed to execute in a Spark cluster only. The available benchmarks state that training/running this framework is faster than the most popular Python frameworks, such as TensorFlow or Caffe, because BigDL uses the Intel Math Kernel Library ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access